Embryonic and fetal beta-globin gene repression by the orphan nuclear receptors, TR2 and TR4

Abstract

The TR2 and TR4 orphan nuclear receptors comprise the DNA-binding core of direct repeat erythroid definitive, a protein complex that binds to direct repeat elements in the embryonic and fetal beta-type globin gene promoters. Silencing of both the embryonic and fetal beta-type globin genes is delayed in definitive erythroid cells of Tr2 and Tr4 null mutant mice, whereas in transgenic mice that express dominant-negative TR4 (dnTR4), human embryonic epsilon-globin is activated in primitive and definitive erythroid cells. In contrast, human fetal gamma-globin is activated by dnTR4 only in definitive, but not in primitive, erythroid cells, implicating TR2/TR4 as a stage-selective repressor. Forced expression of wild-type TR2 and TR4 leads to precocious repression of epsilon-globin, but in contrast to induction of gamma-globin in definitive erythroid cells. These temporally specific, gene-selective alterations in epsilon- and gamma-globin gene expression by gain and loss of TR2/TR4 function provide the first genetic evidence for a role for these nuclear receptors in sequential, gene-autonomous silencing of the epsilon- and gamma-globin genes during development, and suggest that their differential utilization controls stage-specific repression of the human epsilon- and gamma-globin genes.

Figures

Figure 1

DRED binding to the DR Elements of human and mouse embryonic and fetal β-type globin genes. (A) Alignment of the promoter sequences of human and mouse β-type globin gene orthologues. Nucleotides in potential DR elements (horizontal arrows) are shown in bold letters, where as those matching the consensus sequence for nuclear receptor binding are indicated in uppercase. The numbers adjacent to each potential DR element represent the nM Ki determined for that binding site. (B) EMSA competitive binding assays using 1.1 nM 32P-labeled ɛ distal DR probe, and 20 or 200 nM (18- or 180-fold molar excess) unlabeled competitor oligonucleotides. The arrowhead indicates the position of authentic DRED complex. The relative abundance of bound probe is shown at the bottom of each lane (the bound probe with no added competitor set at 100%). −, no competitor. (C) A 10 μg portion of pEF-BOS expression vector driving Flag-tagged TR2 or TR4 cDNA (Tanabe et al, 2002) was transfected separately or together into 293T cells for nuclear extract preparation and EMSA (top panel) or Western blotting with anti-Flag, anti-TR2, or anti-TR4 antibodies (lower panels). The nuclear extract of TR2/TR4 cotransfectant was preincubated with anti-TR2 or anti-TR4 antibody, or preimmune serum, and then subjected to EMSA (rightmost three lanes). A 10 μg portion of a CMV expression vector driving transcription of putative dnTR4 mutant (Flag-tagged) was also transfected into 293T cells. The arrowhead indicates the mobility of the authentic DRED complex from MEL cell nuclear extract. The relative abundance of bound probe is shown at the bottom of each lane (with TR2/TR4 cotransfection set at 100%).

Figure 2

Mouse and human β-type globin gene expression in Tr2 or Tr4 null mutant mice. (A) The abundance of mRNAs for mouse ɛy-, βh1-, and adult β-globin normalized to α-globin mRNA abundance in the 13.5 d.p.c. fetal liver of Tr2 or Tr4 null mutant fetuses and their wild-type littermates was determined by semiquantitative RT–PCR and graphically depicted with s.e.m. The number of animals of each genotype analyzed was 3–8. (B) The abundance of human embryonic ɛ-, fetal γ-, and adult β-globin mRNA normalized to mouse α-globin mRNA abundance in the 14.5 d.p.c. fetal livers of compound Tr2/Tr4 null mutant mice bred to a human β-globin YAC transgenic line was determined by semiquantitative RT–PCR and graphically depicted with s.e.m. The number of fetuses of each genotype analyzed was 2–10. (C) Abundance of primary RNA transcripts for human ɛ-, γ-, and β-globin in 14.5 d.p.c. fetal livers of compound Tr2/Tr4 homozygous null mutant fetuses bearing a wild-type β-globin YAC transgene (Tr2–/–:Tr4–/–:TgβYAC) was determined by RT–PCR and normalized to mouse α-globin mRNA abundance. Relative abundance of primary RNA transcripts normalized to wild-type (Tr2+/+:Tr4+/+:TgβYAC, set at 100%) fetuses is graphically depicted with s.e.m. Two mutant and four wild-type fetuses were analyzed. *P<0.05, **P<0.01, ***P<0.001 by t-test.

Figure 3

Generation of transgenic mice expressing wild-type and dominant-negative TR2 or TR4. (A) Alignment of the amino-acid sequences of the DNA-binding domains of TR4 and RXRα. Mutations introduced to the putative dnTR4 mutant are shown at the top. Residues in RXRα that make base or phosphate contacts are indicated by triangles (Zhao et al, 2000). (B) A 4 μg portion of a CMV expression vector bearing wild-type TR2 (left panel) or TR4 (right) cDNAs was transfected into 293T cells with or without cotransfection of 4 or 16 μg of a CMV expression vector bearing the putative dnTR4, followed by nuclear extract preparation and EMSA. The relative abundance of DR probe bound to wild-type TR2 or TR4 is indicated at the bottom of each lane (bound probe in the absence of dnTR4 was set at 100%). (C) Relative mRNA abundance of the TR2 or TR4 transgenes. The abundance of endogenous (open bars) or transgenic (shaded) TR2 (upper panel) and TR4 (lower) mRNAs in 14.5 d.p.c. fetal livers of transgenic mice expressing dnTR4, or wild-type TR2, TR4 or both was determined by reverse transcription followed by real-time quantitative PCR, and normalized to the abundance of endogenous GATA-1 mRNA (set at 100%). Data represent the averages with s.e.m. of 2–3 fetuses from each transgenic line, or 14 wild-type fetuses.

Figure 4

Human β-globin gene transcription in dnTR4 transgenic mice. (A) The abundance of the human ɛ- and γ-globin mRNAs normalized to the mouse endogenous α mRNA abundance in the 8.5–12.5 d.p.c. yolk sac of transgenic mice bearing the wild-type human β-globin YAC, with or without intercrossed TgdnTR4, was determined by semiquantitative RT–PCR and graphically depicted with s.e.m. One to five embryos of each genotype were examined. (B) The abundance of the human ɛ-, γ-, and β-globin mRNAs normalized to the abundance of α-globin mRNA in 14.5 d.p.c. fetal livers of transgenic mice bearing the wild-type human β-globin YAC, with (+) or without (−) the intercrossed TgdnTR4, was determined. The number of fetuses of each genotype analyzed was 2–7. *P<0.05, **P<0.01 by t-test.

Figure 5

Mouse β-type globin gene expression in transgenic mice forcibly expressing TR2 or TR4. The abundance of mRNAs for mouse embryonic ɛy-, βh1-, and adult β-globin normalized to the mouse endogenous α mRNA abundance in the 9.5 d.p.c. yolk sacs and 14.5 d.p.c. fetal livers of TR2 or TR4 transgenic fetuses and their wild-type littermates was determined by semiquantitative RT–PCR and graphically depicted with s.e.m. The number of animals of each genotype analyzed was 2–6. *P<0.05, **P<0.01, ***P<0.001 by t-test.

Figure 6

Time course of mouse β-type globin mRNA accumulation in TR2/TR4 transgenic mice. The abundance of mRNAs for mouse embryonic ɛy-, βh1-, and adult β-globin normalized to mouse α mRNA in the yolk sac and fetal liver of TR2/TR4 transgenic mice (line 1) and their wild-type littermates from 9.5 to 15.5 d.p.c. was determined by semiquantitative RT–PCR and graphically depicted with s.e.m. Note that the scales for the ɛy- and βh1-mRNA accumulation in the fetal liver are different from the others. Either two or three animals of each genotype were examined. *P<0.05, **P<0.01 by t-test.

Figure 7

Altered human β-type globin gene transcription in TR2/TR4 transgenic mice. (A) The abundance of human ɛ-, γ-, and β-globin mRNAs normalized to mouse α mRNA in the 10.5 d.p.c. yolk sac, 15.5 d.p.c. fetal liver, or adult spleen of transgenic mice bearing a wild-type human β-globin YAC transgene with (+) or without (−) TgTR2/TR4 was determined by semiquantitative RT–PCR and graphically depicted with s.e.m. Two to five animals of each genotype were examined. (B) The abundance of primary RNA transcripts for the same samples as in panel (A) was determined by semiquantitative RT–PCR and normalized to mouse α mRNA abundance. Three animals of each genotype were analyzed. (C) The abundance of human embryonic ɛ-globin mRNA in the 10.5 d.p.c. yolk sac of transgenic mice bearing the DR mutant human β-globin YAC transgene, Bepsi (Tanimoto et al, 1999), in the presence (+) or absence (−) of TgTR2/TR4 (line 2) was determined as described in panel (A). Three fetuses of each genotype were examined. (D) γ-Globin cDNAs from the 15.5 d.p.c. fetal liver of transgenic mice bearing a wild-type or mutDR (Omori et al, 2005) human β-globin YAC transgene either in the presence (+) or absence (−) of TgTR2/TR4 (line 2) were amplified by PCR as in panel (A) and then digested with PstI to determine the Gγ to Aγ molar ratio (Tanimoto et al, 1999; Omori et al, 2005). The averages with s.e.m. for Gγ- and Aγ-globin mRNAs normalized to mouse endogenous α-globin are graphically depicted. Three fetuses of each genotype were examined. *P<0.05, **P<0.01, ***P<0.001 by t-test.

Figure 8

Time course of human β-type globin mRNA accumulation in TR2/TR4 transgenic mice. The abundance of the human ɛ-, γ-, and β-globin mRNAs normalized to mouse α-globin mRNA in the yolk sac and fetal liver of transgenic mice bearing the wild-type human β-globin YAC, with or without TgTR2/TR4 (line 1), from 9.5 to 15.5 d.p.c. was determined by semiquantitative RT–PCR and graphically depicted with s.e.m. Note that the scales for ɛ- and γ-globin mRNA accumulation in the fetal liver are different from the others. Two to five animals of each genotype were used. *P<0.05, **P<0.01, ***P<0.001 by t-test.

Figure 9

A model for the role of TR2/TR4 in developmental stage-specific silencing of the human ɛ- and γ-globin genes. In primitive erythroid cells (top), DRED is formed as a complex of TR2/TR4 and other (currently unidentified) co-repressors, and DRED represses ɛ-globin transcription, but exerts little or no effect on the γ-globin gene in primitive erythroid cells (because of higher affinity for the ɛ-globin promoter DR sites, and/or because of its activity/abundance at that stage). In definitive erythroid cells (models 1 and 2), the activity of DRED increases, allowing it to gradually repress γ-globin synthesis from the (lower affinity) DR site in the γ-globin promoter. The seemingly contradictory induction of the γ-globin genes in the TgTR2/TR4 transgenic mice may be explained either by a dominant-negative effect of the forcibly expressed wild-type receptors which dilute limiting γ-specific co-repressors for their normal repressor activity against the γ-globin gene (model 1), or by an inherent context-dependent transcriptional activator function (i.e. dual functionality) of the receptors on the γ-globin gene, depending on interaction with specific coactivators that are found only in definitive erythroid cells (model 2).